✦ScienexySCIENCE KNOWLEDGE LIBRARY
reference · SCI-REF-003-077

Reproducibility and Intermediate Precision

Reproducibility and intermediate precision describe measurement precision under conditions that vary beyond those used for repeatability. Intermediate precision evaluates variation within a laboratory when specified conditions such as operators, days or equipment change. Reproducibility evaluates precision under reproducibility conditions, often involving different laboratories, operators, equipment or measurement systems. These concepts are important for determining whether measurement methods produce consistent results across broader operating conditions.

For a simple nested measurement model:
xᵢⱼₖ = μ + Lᵢ + Oᵢⱼ + εᵢⱼₖ
where the model represents a general hierarchy of laboratory, operator or condition, and replicate contributions.
Total variance for independent components: σ²_total = σ²_L + σ²_O + σ²_r.
Combined standard deviation: s_total = √(s²_L + s²_O + s²_r).
For a simple two-level random-effects model, variance components may be estimated using ANOVA mean squares, but the exact equations depend on the study design and number of replicates.

Explanation

A measurement method may perform consistently when one operator uses one instrument in one laboratory, yet produce different results when used by another operator, on another day or in another laboratory. Repeatability alone cannot describe this broader performance. Intermediate precision and reproducibility provide additional information about variation when measurement conditions change.

Intermediate precision is measurement precision under a set of intermediate precision conditions within a single laboratory or measurement environment. These conditions may include changes in measurement day, operator, equipment, calibration status or other specified factors. Not every factor must change simultaneously. The study should identify which conditions are varied and which are kept controlled.

Reproducibility is measurement precision under reproducibility conditions. In many measurement studies, this involves different laboratories, operators, measuring systems, equipment or other specified conditions. Reproducibility studies help establish whether a measurement method can produce comparable results when transferred between independent locations or organizations.

The three precision conditions can be viewed as progressively broader measurement environments. Repeatability focuses on closely controlled conditions. Intermediate precision examines selected changes within a laboratory. Reproducibility examines a wider set of changes, frequently across laboratories. These conditions are not simply levels of instrument quality; they describe the circumstances under which measurement variation is assessed.

Variation can arise from several sources. Within one laboratory, different operators may position a sample differently or interpret a procedure differently. Equipment may have different calibration characteristics. Environmental conditions may change between days. Between laboratories, differences in reference standards, instrument models, sample preparation, environmental control and procedural interpretation may contribute to additional variation.

A well-designed precision study uses a clear protocol, representative samples, appropriate replicate measurements and a statistical analysis suited to the experimental design. The study should record the factors that change and avoid confounding them unnecessarily. For example, if the operator, instrument and day all change together, it may be difficult to identify which factor contributed most to the observed variation.

Analysis of variance (ANOVA) and other suitable statistical methods can separate within-condition and between-condition variation when the experimental design supports such analysis. For a simple balanced nested design, variance components can be estimated for repeatability and higher-level sources. More complex designs may require mixed-effects models or other methods.

Reproducibility does not mean that all laboratories must obtain numerically identical results. It means that the precision of results under the specified reproducibility conditions has been evaluated. Differences between laboratory means may also reveal bias or method implementation differences, which should be examined separately from precision.

Variables

SymbolMeaningUnit

xᵢⱼₖ

The kth measured result associated with the specified higher-level conditions

Unit of the measurand

μ

Overall mean or underlying central value in the simplified model

Unit of the measurand

Lᵢ

Contribution associated with laboratory or another highest-level condition

Unit of the measurand

Oᵢⱼ

Contribution associated with operator or an intermediate condition

Unit of the measurand

εᵢⱼₖ

Residual or repeatability contribution

Unit of the measurand

σ²_total

Total variance under the specified model

Square of the measurand unit

σ²_L

Variance component associated with the laboratory or highest-level factor

Square of the measurand unit

σ²_O

Variance component associated with the intermediate factor

Square of the measurand unit

σ²_r

Repeatability variance component

Square of the measurand unit

Example or application

A worked example has not yet been added.

Definition

Repeatability

Measurement precision under specified repeatability conditions.

Intermediate Precision

Measurement precision under a set of intermediate precision conditions within a laboratory or measurement environment, where selected factors such as time, operator or equipment may vary.

Reproducibility

Measurement precision under specified reproducibility conditions, which commonly involve different laboratories, operators or measuring systems.

Purpose

The purpose of evaluating intermediate precision and reproducibility is to understand how measurement results behave when selected measurement conditions change, and to determine whether a method is sufficiently consistent for routine use, transfer or comparison across laboratories.

Importance

These characteristics are important for method validation, laboratory accreditation, interlaboratory comparisons, manufacturing quality assurance and scientific collaboration. They reveal sources of variation that may remain hidden in a single-operator repeatability study.

Precision Conditions

ConditionEnvironmentVariation Assessed

Repeatability

Same procedure, operator, measuring system, location and substantially unchanged operating conditions over a short period.

Variation under closely controlled conditions.

Intermediate precision

Same laboratory, with selected factors such as operator, day, equipment or calibration changing.

Within-laboratory variation under specified intermediate conditions.

Reproducibility

Different laboratories or other broadly changed measurement conditions as defined by the study.

Variation under the specified reproducibility conditions.

Factors Affecting Precision

FactorExplanation

Operator

Different operators may interpret instructions, prepare samples or position instruments differently.

Measurement day

Environmental conditions, instrument stability and laboratory operations may change over time.

Equipment

Different instruments may have different response characteristics, resolution and calibration histories.

Laboratory

Facilities, reference standards, environmental control and working practices may differ.

Sample preparation

Different preparation procedures can introduce variation before the actual measurement.

Method interpretation

Ambiguous instructions can cause laboratories or operators to implement different procedures.

Study Design

Steps

  • Define the measurand, measurement method and intended use.

  • Specify the repeatability, intermediate precision or reproducibility conditions to be studied.

  • Select representative samples covering the relevant measurement range.

  • Plan the number of operators, days, instruments, laboratories and replicate measurements.

  • Keep the experimental design balanced where practical and record every relevant condition.

  • Perform measurements using a controlled and documented procedure.

  • Check the data for recording mistakes, unsuitable observations and unexpected patterns.

  • Use appropriate statistical methods to estimate variation and variance components.

  • Report the study design, estimated precision and limitations.

Important Consideration

The study design determines what can be concluded. If several factors change together, the study may estimate overall variation but may not identify the separate contribution of each factor.

Statistical Methods

MethodUse

Standard deviation

Summarizes the spread of repeated observations under a defined set of conditions.

Analysis of variance

Can separate variance components associated with different factors when the design and statistical assumptions are suitable.

Mixed-effects models

Can model fixed and random effects in more complex measurement designs.

Control charts

Can monitor measurement-process stability over time and identify changes that require investigation.

Interlaboratory comparison

Compares results from different laboratories using an agreed sample or measurement procedure.

Worked Examples

TitleProblemSolutionInterpretation

Repeatability compared with intermediate precision

An operator measures a reference sample five times using the same instrument in one session. Later, three operators measure the same sample on different days.

The first group evaluates repeatability under the stated conditions. The second group can contribute to an intermediate precision study because operator and day vary within the laboratory.

The second study examines a wider range of conditions than the first.

Understanding reproducibility

Three independent laboratories use the same documented method to measure the concentration of a reference material.

The variation among results under the defined interlaboratory conditions contributes to the evaluation of reproducibility. Any differences in laboratory means should also be investigated for possible bias or method implementation differences.

Reproducibility describes precision across the specified laboratory conditions; it does not automatically prove that every laboratory is unbiased.

Combining independent variance components

A simplified study estimates repeatability standard deviation as 0.10 units and an independent between-condition standard deviation component as 0.20 units.

s_total = √(0.10² + 0.20²) = √0.05 ≈ 0.224 units.

The combined variation is larger than the repeatability variation alone. This calculation is valid only when the components are appropriately estimated and independent under the model.

Common Misconceptions

  • Reproducibility is not simply another word for repeatability.

  • Intermediate precision is not necessarily a study in which every possible factor changes.

  • Reproducibility does not require identical numerical results from all laboratories.

  • A reproducibility study can reveal variation but does not by itself establish the cause of that variation.

  • A method can have good repeatability but poor reproducibility.

  • The number and selection of laboratories, operators and samples influence the conclusions of a precision study.

Applications

  • Analytical method validation

  • Interlaboratory comparisons

  • Calibration and testing laboratories

  • Medical laboratory quality assurance

  • Pharmaceutical and chemical analysis

  • Manufacturing inspection

  • Environmental testing

  • Research method transfer and standardization

Related Topics

  • Accuracy, Precision, Trueness and Repeatability

  • Systematic, Random and Gross Errors

  • Measurement Uncertainty

  • Calibration and Traceability

  • Statistical Analysis of Measurement Data

Reference

TitleOrganizationUrl

International Vocabulary of Metrology (VIM), JCGM 200

Joint Committee for Guides in Metrology

https://www.bipm.org/en/committees/jc/jcgm/publications

ISO 5725 — Accuracy (Trueness and Precision) of Measurement Methods and Results

International Organization for Standardization

https://www.iso.org/

NIST/SEMATECH e-Handbook of Statistical Methods

National Institute of Standards and Technology

https://www.itl.nist.gov/div898/handbook/

Related Topic

  • SCI-REF-003-074

  • SCI-REF-003-076

  • SCI-REF-003-078

  • SCI-REF-003-080

Sources and verification

Sources are shown as provided in the project data. Not every source has yet been independently verified. Please report errors or missing citations.

  • Source verification pending; check an authoritative textbook, university, standards body, or peer-reviewed source.

Suggest a correction or source · More in Units & Measurement